IRO Journals
Journal of Innovative Image Processing

An Optimized Deep Learning Framework for Fetal Plane Classification using Hybrid Attention-Based Models

Open Access
Published: 23 October, 2026
Pages: 1557-1581
  • Harismithaa L R. , Harismithaa L R.
    Research Scholar, Department of Computer Science and Engineering, PSG College of Technology, Coimbatore, India
    Research Scholar, Department of Computer Science and Engineering, PSG College of Technology, Coimbatore, India
  • Sudha Sadasivam G. Sudha Sadasivam G.
    Professor, Department of Computer Science and Engineering, PSG College of Technology, Coimbatore, India
    Professor, Department of Computer Science and Engineering, PSG College of Technology, Coimbatore, India
view PDF
view PDF

How to Cite

L R., Harismithaa, and Sudha Sadasivam G. 2026. “An Optimized Deep Learning Framework for Fetal Plane Classification Using Hybrid Attention-Based Models”. Journal of Innovative Image Processing 8 (4): 1557-81. https://doi.org/10.36548/jiip.2026.4.011.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Attention
Convolutional Block Attention Module (CBAM)
Depth-wise Separable Convolutions
Fetal Plane
Ghost Module
Ultrasound Imaging

Abstract

Fetal plane classification is the foremost step in analyzing fetal health in ultrasound screening. Standard planes from fetal ultrasound scans are used to estimate fetal biometric parameters, such as femur length and overall fetal growth. This work develops an automated method for identifying standard and non-standard fetal anatomical planes in 2D ultrasound images to assist sonographers and reduce operator workload and variability. A pre-trained neural network is selected and trained for the fetal anatomy classification task. The trained network is integrated with the convolutional block attention module to improve the classification and localization of important features in ultrasound images. Further, to minimize computational demands, an efficiency-aware integration is proposed, in which a Ghost module offsets the parameter and FLOP overhead of CBAM within an EfficientNet-B0 backbone. The results have shown improved performance, as measured by detailed metrics, with a 15% reduction in trainable parameters and a 7% reduction in FLOPs relative to the attention-only model. A detailed class-wise analysis of sensitivity and specificity highlights the efficiency of the proposed model in complex fetal planes. The synergistic deep learning framework for automated fetal standard and non-standard ultrasound plane classification using a hybrid attention mechanism exhibits notably good computational performance and improved robustness compared to existing literature. The proposed model was validated on a four-class test split of the BCNatal cohort and an independent multicentre cohort acquired in five African countries. In addition, a four-way ablation study is presented to evaluate the performance and computational efficiency on GPU and resource-constrained hardware. The proposed learning framework strives to achieve a balance between computational overhead and classification efficiency within a lightweight framework for the fetal plane classification task.

References

  1. Prabakaran, Bharath Srinivas, Paul Hamelmann, Erik Ostrowski, and Muhammad Shafique. “FPUS23: An Ultrasound Fetus Phantom Dataset with Deep Neural Network Evaluations for Fetus Orientations, Fetal Planes, and Anatomical Features.” IEEE Access 11 (2023): 58308-58317.
  2. Burgos-Artizzu, Xavier P., David Coronado-Gutiérrez, Brenda Valenzuela-Alcaraz, Elisenda Bonet-Carne, Elisenda Eixarch, Fatima Crispi, and Eduard Gratacós. “Evaluation of Deep Convolutional Neural Networks for Automatic Classification of Common Maternal Fetal Ultrasound Planes.” Scientific Reports 10, no. 1 (2020): 10200.
  3. Harikumar, Akshay, Simi Surendran, and S. Gargi. “Explainable AI in Deep Learning Based Classification of Fetal Ultrasound Image Planes.” Procedia Computer Science 233 (2024): 1023-1033.
  4. Srivastava, Somya, Ankit Vidyarthi, and Shikha Jain. “A Highly Densed Deep Neural Architecture for Classification of the Multi-Organs in Fetal Ultrasound Scans.” Neural Computing and Applications 37, no. 28 (2025): 23619-23633.
  5. Krishna, Thunakala Bala, and Priyanka Kokil. “Automated Classification of Common Maternal Fetal Ultrasound Planes Using Multi-Layer Perceptron with Deep Feature Integration.” Biomedical Signal Processing and Control 86 (2023): 105283.
  6. Ghabri, Haifa, Mohammed S. Alqahtani, Soufiene Ben Othman, Amal Al-Rasheed, Mohamed Abbas, Hassan Ali Almubarak, Hedi Sakli, and Mohamed Naceur Abdelkarim. “Transfer Learning for Accurate Fetal Organ Classification from Ultrasound Images: A Potential Tool for Maternal Healthcare Providers.” Scientific Reports 13, no. 1 (2023): 17904.
  7. Tan, Mingxing, and Quoc Le. “Efficientnet: Rethinking Model Scaling for Convolutional Neural Networks.” In International conference on machine learning, PmLR, 2019, 6105-6114.
  8. Sendra-Balcells, Carla, Víctor M. Campello, Jordina Torrents-Barrena, Yahya Ali Ahmed, Mustafa Elattar, Benard Ohene-Botwe, Pempho Nyangulu et al. “Generalisability of Fetal Ultrasound Deep Learning Models to Low-Resource Imaging Settings in Five African Countries.” Scientific reports 13, no. 1 (2023): 2728.
  9. Govardhan, C., and V. Anantha Natarajan. “Interpretable and Efficient Brain Tumor Detection using Efficient Net Enhanced with CBAM and Grad-CAM Analysis.” International Journal of Artificial Intelligence and Machine Learning 6, no. 1s (2026): 487-505.
  10. Li, Junjie, Weiming Mi, Yucheng Guo, Xinyu Ren, Hao Fu, Tao Zhang, Hao Zou, and Zhiyong Liang. “Artificial Intelligence for Histological Subtype Classification of Breast Cancer: Combining Multi‐Scale Feature Maps and the Recurrent Attention Model.” Histopathology 80, no. 5 (2022): 836-846.
  11. Sivasubramanian, Arrun, Divya Sasidharan, V. Sowmya, and Vinayakumar Ravi. “Efficient Feature Extraction Using Light-Weight CNN Attention-Based Deep Learning Architectures for Ultrasound Fetal Plane Classification.” Physical and Engineering Sciences in Medicine 48, no. 3 (2025): 1079-1093.
  12. Ogut, Zeki, Mucahit Karaduman, and Muhammed Yildirim. “Clinically Focused Computer-Aided Diagnosis for Breast Cancer Using Se and Cbam with Multi-Head Attention.” Tomography 11, no. 12 (2025): 138.
  13. Omega Boro, Lal, and Gypsy Nandi. “CBAM-RIUnet: Breast Tumor Segmentation with Enhanced Breast Ultrasound and Test-Time Augmentation.” Ultrasonic Imaging 47, no. 1 (2025): 24-36.
  14. Shih, Yi-Hsuan, Chih-Ying Wu, Ruei-Chi Lin, and Cheng-Ta Huang. “Segmentation of Tumor Regions in BUSI Breast Ultrasound Images based on DRA-UNet Model with CBAM.” In Proceedings of the 2025 7th international conference on image, video and signal processing, pp. 105-113. 2025.
  15. Huai, Ximing, Lei Jiang, Chao Wang, Peng Chen, and Hanchi Li. “Heart Sound Classification Based on Convolutional Neural Network with Convolutional Block Attention Module.” Frontiers in Physiology 16 (2025): 1596150.
  16. Jaba Deva Krupa, Abel, Samiappan Dhanalakshmi, Khin Wee Lai, Yongqi Tan, and Xiang Wu. “An IoMT Enabled Deep Learning Framework for Automatic Detection of Fetal QRS: A Solution to Remote Prenatal Care.” Journal of King Saud University Computer and Information Sciences 34, no. 9 (2022): 7200-7211.
  17. Teoh, Jing Ru, Khairunnisa Hasikin, Khin Wee Lai, Xiang Wu, and Chong Li. “Enhancing Early Breast Cancer Diagnosis Through Automated Microcalcification Detection Using an Optimized Ensemble Deep Learning Framework.” PeerJ Computer Science 10 (2024): e2082.
  18. Lin, Yirong, Xu Guo, and Peizhong Liu. “FBSPM-Net: Automatic Recognition of Fetal Brain Ultrasound Images in Early Pregnancy.” In Proceedings of the 2025 International Conference on Health Big Data, 2025, 80-85.
  19. LR, Harismithaa, and G. Sudha Sadasivam. “An Intensive Review on Deep Learning Approaches in Non-Invasive Fetal Healthcare Monitoring.” In 2024 4th International Conference on Sustainable Expert Systems (ICSES), IEEE, 2024, 670-676.
  20. Han, Kai, Yunhe Wang, Qi Tian, Jianyuan Guo, Chunjing Xu, and Chang Xu. “Ghostnet: More Features from Cheap Operations.” In 2020 IEEE/CVF conference on computer vision and pattern recognition (CVPR), IEEE, 2020, 1577-1586.
  21. Chen, Long, Hanwang Zhang, Jun Xiao, Liqiang Nie, Jian Shao, Wei Liu, and Tat-Seng Chua. “Sca-cnn: Spatial and Channel-Wise Attention in Convolutional Networks for Image Captioning.” In 2017 IEEE conference on computer vision and pattern recognition (CVPR), IEEE, 2017, 6298-6306.
  22. Rauf, Fatima, Muhammad Attique Khan, Hussain M. Albarakati, Kiran Jabeen, Shrooq Alsenan, Ameer Hamza, Sokea Teng, and Yunyoung Nam. “Artificial Intelligence Assisted Common Maternal Fetal Planes Prediction from Ultrasound Images Based on Information Fusion of Customized Convolutional Neural Networks.” Frontiers in Medicine 11 (2024): 1486995.
  23. Wu, Zizhang, Man Wang, Weiwei Sun, Yuchen Li, Tianhao Xu, Fan Wang, and Keke Huang “CAT: Learning to Collaborate Channel and Spatial Attention from Multi‐Information Fusion.” IET Computer Vision 17, no. 3 (2023): 309-318.
  24. Sengodan, Naren. “Breast Cancer Histopathology Classification Using CBAM-EfficientNetV2 with Transfer Learning.” arXiv preprint arXiv:2410.22392 (2024).
  25. Woo, Sanghyun, Jongchan Park, Joon-Young Lee, and In So Kweon. “Cbam: Convolutional Block Attention Module.” In European conference on computer vision, Cham: Springer International Publishing, 2018, 3-19.